It has long been recognized that neuroanatomic abnormalities are unlikely to be the sole cause of brain dysfunction in schizophrenia. Neuroimaging methods like MRI and PET have provided an opportunity for detailed in vivo anatomical and functional investigations of schizophrenia. In spite of all these studies, there have been relatively few integrating hypotheses put forward for the pathogenesis of this disease. The emergence of localized Magnetic Resonance Spectroscopy has provided new opportunities for investigating the neurobiochemistry of schizophrenia. Results reported in the literature obtained using localized 31-P MRS have shown altered metabolism in both the frontal and temporal lobes of patients with schizophrenia. The present proposal employs localized proton MRS to study regional metabolic alterations in unmedicated first episode patients (n=60) and compare these results to similar studies carried out on matched normal controls (n=60). The ability to monitor many compounds; e.g. N-acetyl aspartate, which is regarded as a marker of neuronal integrity; neurotransmitters such as glutamate, aspartate and gaba; compounds involved in membrane synthesis such as choline; as well as compounds with other biochemical functions such as myo-inositol enables proton MRS to simultaneously probe many diverse biochemical processes. The results of our localized proton MRS studies of first episode schizophrenics carried out in the initial funding period of this project demonstrate clear metabolic abnormalities in the frontal and temporal lobes; reductions in NAA in the frontal lobe and elevations in amino acids in the temporal lobes. In our funded study, we were able to sample 3.35 cm3 regions of the brain using the convention head coil employed for MRI. We have developed, validated and implemented a high resolution MRS method at 1.5 T using local coils which can selectively sample regions of the brain at 0.36 cm3 resolution. This development makes it possible to begin to probe the metabolic alterations observed in pure cortical gray matter and white matter regions. This method will be incorporated into our existing protocol in order to test various metabolic hypotheses (i.e. frontal lobe hypofunctional hypothesis, synaptic pruning, glutamatergic hypothesis etc.) that have been proposed for schizophrenia in particular in its early stages. The recent installation of a 4T research whole body MR scanner in our department offers significant potential advantages to this study. We propose to assess the advantages of the 4 T scanner over 1.5 T scanners on the same patient cohort.